Visually-augmented pretrained language models for NLP tasks without images

37638

Knowledge Triples

Source Entity	Relationship	Target Entity
Visual Knowledge Augmentation Method	augments	Pre-trained Language Models Method
Visually Augmented Fine Tuning Method	appliesTo	BART Method
Visually Augmented Fine Tuning Method	appliesTo	BERT Method
Visually Augmented Fine Tuning Method	appliesTo	RoBERTa Material
Visually Augmented Fine Tuning Method	appliesTo	T5 Method
Visually Augmented Fine Tuning Method	improves	Pre-trained Language Models Method
Visually Augmented Fine Tuning Method	outperforms	Competitive Baselines OtherEntity